Counterions control whether self-assembly leads to formation of stable and well-defined unilamellar nanotubes or nanoribbons and nanorods

Article abstract of DOI:10.1002/chem.201402523

Self-assembly of the amphiphilic π-conjugated carbenium ion ATOTA-1 ⁺ in aqueous solution selectively leads to discrete and highly stable nanotubes or nanoribbons and nanorods, depending on the nature of the counterion (Cl^- vs. PF₆^-, respectively). The nanotubes formed by the Cl^- salt illustrate an exceptional example of a structural well-defined (29±2 nm in outer diameter) unilamellar tubular morphology featuring π-conjugated functionality and high stability and flexibility, in aqueous solution.

Full text of DOI:10.1002/chem.201402523

DOI: 10.1002/chem.201402523
Communication
&
Amphiphiles
Counterions Control Whether Self-Assembly Leads to Formation
of Stable and Well-Defined Unilamellar Nanotubes or
Nanoribbons and Nanorods
Dong Shi, Christian Schwall, George Sfintes, Erling Thyrhaug, Peter Hammershøj,
Marite Cꢀrdenas, Jens B. Simonsen, and Bo W. Laursen*[a]
In memory of George Sfintes
ture of the counterion of the cationic discotic molecule offers
Abstract: Self-assembly of the amphiphilic p-conjugated
new opportunities to control aggregate structures[3e,5] forming,
carbenium ion ATOTA-1+ in aqueous solution selectively
for example, stable planar aggregates. The supramolecular ag-
leads to discrete and highly stable nanotubes or nanorib-
gregates mentioned above were formed in either THF or 50 to
bons and nanorods, depending on the nature of the
100% methanol solutions. Recently, Shao et al. showed how
counterion (ClÀ vs. PF6À, respectively). The nanotubes
the variation of the hydrophilic headgroup of an amphiphilic
formed by the ClÀ salt illustrate an exceptional example of
discotic leads to the formation of either uniform sized (diame-
a structural well-defined (29Æ2 nm in outer diameter) uni-
ter of ca. 10 nm) unilamellar or multilamellar nanotubes in
aqueous solution.[6] However, the compound forming the uni-
lamellar tubular morphology featuring p-conjugated func-
tionality and high stability and flexibility, in aqueous solu-
lamellar tubes was completely insoluble at neutral pH.
tion.
Herein, we combine p-conjugated discotics, carbocationic
charge, and tunability of counterions to form either uniform
29 nm wide and highly stable unilamellar nanotubes, or multi-
Self-assembly of p-conjugated discotic (disc-like) molecules to
form materials and nanostructures containing columnar p-
stacks is of great interest for supramolecular optoelectronics.
The columnar stacks of p-systems make up a fundamental
supramolecular structure element, and provide a one-dimen-
sional pathway for charge or exciton transport.[1] Discotics
equipped with a symmetric corona of flexible alkyl chains
dominate the field of columnar liquid crystals and have suc-
cessfully been applied in various thin-film devices.[2] However,
asymmetric discotics with alkyl chains on only one side of the
disc and no chains or polar groups on the opposite side of the
rim obtain amphiphilic properties and tend to organize the
columnar aggregates in bilayer structures, which may lead to
nanostructures, such as fibers and tubes.[3] The increased com-
plexity of self-assembled nanotubes, compared to the most
common self-assembled structures, such as micelles, vesicles,
and lyotropic liquid crystals, offers both an academic challenge
and possibilities for more advanced structural and optoelec-
tronic functions.[3b,4] Aida and co-workers have shown that an
amphiphilic variant of the molecular analog to graphite; hexa-
benzocorenene forms nanotubes in organic solution.[3c] Mꢁllen
and co-workers demonstrated that tuning the size and struc-
layer nanoribbons and nanorods in aqueous solution. We dem-
onstrate how the interaction between bilayers of the amphi-
philic cationic discotic dye in aqueous solution can be tuned
from attractive to repulsive by choice of anion, leading to
these distinct and complex nanostructures.
Previously, we reported the synthesis and columnar self-as-
sembly of dodecyl-aminobisdimethylaminotrioxatriangulenium
hexafluoro-phosphate (ATOTA-1·PF6, Figure 1) in Langmuir
films at the air/water interface,[7] and in solid multilayer Lang-
muir–Blodgett (LB) films.[8]
The ATOTA-1·Cl and ATOTA-1·PF6 were synthesized (see the
Supporting Information) to study the three-dimensional self-as-
sembly of this discotic amphiphile in aqueous solution, and
the role of the associated counterions. The ATOTA-1+-based
nanostructures were obtained by dilution of the ATOTA-1+ salt
dissolved in acetonitrile (MeCN) with 10- to 100-fold of water
(see the Supporting Information). The acetonitrile stock solu-
[a] Dr. D. Shi, C. Schwall, G. Sfintes, Dr. E. Thyrhaug, Dr. P. Hammershøj,
Dr. M. Cꢀrdenas, Dr. J. B. Simonsen, Dr. B. W. Laursen
Nano-Science Center & Department of Chemistry
University of Copenhagen, Universitetsparken 5
2100 Copenhagen (Denmark)
E-mail: bwl@nano.ku.dk
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201402523.
Figure 1. Molecular structure of the amphiphilic amino-trioxatriangulenium
À
dye ATOTA-1+; XÀ =PF6 or ClÀ.
Chem. Eur. J. 2014, 20, 1 – 5
1
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Communication
tions (10À3 to 10À4 m) showed the characteristic absorption and
fluorescence spectra of the fully dissolved and non-aggregated
ATOTA + cation.[9] The addition of water resulted in blueshift of
the absorption and redshift of the fluorescence indicating the
presence of columnar co-facial p–p aggregates[7–8] (Figures S1
and S2 in the Supporting Information). The absorption spectra
of the ATOTA-1+ salts in aqueous solution were concentration
independent in the measured range from 10À4 to 10À7 m, and
showed no variation with the acetonitrile content between
1 to 10 vol%. The data described below are based on the
10 vol% MeCN aqueous solution. No significant differences in
absorption and fluorescence were found between the aggre-
close inspection of the ribbons revealed in several cases regu-
lar fine structures at the edge as shown by the cross-section in
Figure 2C. An equal numbers of layers were observed on both
sides of the ribbon, indicating that the ribbon is a closed struc-
ture, rather than just an elongated assembly of layers. A uni-
form repeated thickness of 40 ꢃ was found for all the observed
layers. For the larger twisting rods (2) similar 40 ꢃ layers oc-
curred across the entire width of the rod (Figure 2D). The
cross-section (Figure 2D and E) and the fast Fourier transform
analysis (Figure S5 in the Supporting Information) confirmed
that the 40 ꢃ layer is a fundamental structural element in both
the ribbons and the rods. In addition, atomic force microscopy
(AFM) and X-ray diffraction studies of dried samples showed
similar features (Figures S6 and S7 in the Supporting Informa-
tion). AFM analysis of the twisting rods (2) showed uniform
kink angles and periodicity along the rods (Figures S8–S10 in
the Supporting Information), leading to the conclusion that
the each twisting rod is made up by a single ribbon (1) twist-
ing around itself.
À
gates formed by ClÀ and PF6 salts. Dynamic light scattering
(DLS) showed in both cases two modes with no linear depend-
ence with the scattering angle, typical for large and elongated
aggregates (Figure S3 in the Supporting Information).[10]
The ATOTA-1·PF6 (10À4 m) aggregates in water (with 10 vol%
MeCN) were studied by cryo-transmission electron microscopy
(cryo-TEM). Two types of nanostructures were observed (Fig-
ure 2A). The dominant structure was 60 to 100 nm wide nano-
ribbons (1), whereas larger periodical twisting rods (2) with a di-
ameter of approximately 150 nm were also present. It was diffi-
cult to determine the exact length of these structures. Howev-
er, nanorods of several microns were in general observed (Fig-
ure S4 in the Supporting Information).
The largest observed width of the untwisted ribbons (1) is
100–120 nm, indicating that the folding into rods (2) occurs as
the ribbons reach a certain critical size, close to 100 nm. Twist-
ing super-structures are observed for many self-assembled
nanoribbons and fibers and may be driven by internal strain
and solvophobic interactions.[11] Wide-angle X-ray scattering
(WAXS) from an aqueous solution (10 vol% MeCN) of ATOTA-
1·PF6 showed two pronounced Bragg peaks corresponding to
a repeat distances of 3.5 and 39.2 ꢃ (Figure 2F). The 39.2 ꢃ
peak confirms the lamella structure observed by TEM and
AFM, whereas the 3.5 ꢃ corresponds to a characteristic co-
facial p–p packing distance and confirms that a columnar p-
stacking motif is certainly is present in these multilayer nano-
structures. A smaller peak at approximately 3.3 ꢃ (also ob-
served in the dry samples, Figure S7 in the Supporting Infor-
mation) might be an indication of the co-existence of a more
densely packed and crystalline p-stack,[12] for example, in the
twisting rods.
The nanoribbons were characterized by uniform density and
straight edges, except for areas, in which occasional twists of
the ribbons occurred (Figure 2B). The flat-ribbon structure
could indicate a simple lamella packing of layers. However,
Although UV/Vis spectroscopy and DLS did not provide
À
proof of any significant differences between the PF6 and the
ClÀ salts, cryo-TEM data indeed did. Figure 3 shows cryo-TEM
micrographs of a 10À4 m solution of ATOTA-1·Cl prepared in the
Figure 2. Cryo-TEM micrographs (A–E) and WAXS data (F) of ATOTA-1·PF6
nanostructures formed in water (10 vol% MeCN). A) Nanoribbons labeled 1,
and twisted nanorods labeled 2. B) Single kink on nanoribbon. C) Cross-sec-
tion of nanoribbon in B. D) Enlarged twisted nanorod. E) Cross-section analy-
sis of the gray scale along the yellow line in D. F) The presented data is
background subtracted.
Figure 3. Cryo-TEM micrographs of ATOTA-1·Cl aggregates. A),B) ATOTA-1·Cl
(10À4 m) in water (10 vol% MeCN). C) ATOTA-1·Cl (10À4 m) in pure water (vor-
texed).
&
&
Chem. Eur. J. 2014, 20, 1 – 5
www.chemeurj.org
2
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Communication
À
same way as the solutions of the PF6 salt. Contrary to the
À
PF6 salt, the ClÀ salt of ATOTA-1+ formed unilamellar tubes.
The tubes showed an extremely uniform width of 29Æ2 nm
and a wall thickness of approximately 4 nm, corresponding
very well to the bilayer dimension found in ATOTA-1·PF6 LB
films,[8] and the layered structure observed in the rods and rib-
À
bons of the PF6 salt. The high similarity of the optical features
and identical bilayer arrangement between the nanoribbons/
rods and the nanotubes confirmed that a similar p-stacking
motif is present in the tubes. The tubes differ from the relative-
Figure 4. Schematic outline of the molecular dimensions of the ATOTA-1+
amphiphile (A), the columnar stack (B) and cross-section of the bilayer (C).
À
ly rigid PF6 aggregates by showing a high flexibility, for exam-
ple, by making a full 1808 turn in less than 100 nm (Figure 3A).
Interestingly, we were also able to form the 29 nm nano-
tubes by vigorously shaking ATOTA-1·Cl powder in pure water
for 24 h at 408C (Figure 3C). However, in this case, the most
abundant feature was a twisting nanoribbon with a contrast in
the range expected for a single bilayer wall of the tube struc-
tures. Several groups have shown that twisted ribbons are
a potential precursor to formation of supramolecular nano-
tubes,[13] including some amphiphilic p-conjugated discotics.[6]
The shaking experiment in 100% water indicates that MeCN
during the dilution with water is promoting nanotube forma-
tion rather than playing a crucial role in stabilizing the tubular
morphology in the water based solutions (1–10 vol% MeCN).
The nanotubes do not agglomerate, fuse, or form multila-
mellars over time, and are well-separated, even in regions with
a high density of tubes. The long-term stability of the tubes
was also indirectly confirmed by showing no sign of precipita-
tion upon storage for months at ambient conditions. In con-
trast, the ATOTA-1·PF6 aggregates precipitated over time in
aqueous solution. Precipitation is a common behavior of p-
conjugated amphiphilic discotics that form nanotube, rod, and
ribbon morphologies in organic-based solutions.[3c,14]
tubes are not assigned. Nevertheless, the co-existence of twist-
ed nanoribbons and tubes in the sample prepared by shaking
ATOTA-1·Cl powder in pure water (Figure 3C), indicates that
these single bilayer nanoribbons may be precursors for the
tubes.[13]
In the case of the ATOTA-1·PF6 salt, the multilayer ribbons
have a structure that in principle can be constructed by wrap-
ping multiple bilayers around a flat ribbon or a flattened/col-
lapsed tube. The twisting rods are assigned as higher-order
structures formed by twisting/folding of the multilayer rib-
bons.
Although the internal p-stacking columnar bilayer structure
À
is proposed to be the same for the ClÀ and PF6 aggregates,
two major differences are observed: 1) The ClÀ salt forms strict-
À
ly unilamellar tubes, and the PF6 salt—solid multilamellar
À
structures; 2) The PF6 nanoribbons show size distribution (60–
100 nm), form higher-order aggregates and precipitates over
time, whereas the ClÀ tubes are very uniform (29Æ2 nm),
highly flexible, and stable in aqueous solution.
We propose that these significant counterion effects largely
can be explained by the differences in solvent interactions and
the resulting Coulomb interactions between the ionic super-
structures. Although the outer surface of the ATOTA-1+ bilay-
ers have a relative high charge density (+1 e/ꢀ60 ꢃ2), the
ATOTA + ions are large and extremely polarizable,[9a,12,16] inter-
A transition from the unilamellar ClÀ nanotubes to the multi-
lamellar morphologies could be induced by adding equal
molar amounts of KPF6 to a stabile solution of ATOTA-1·Cl
nanotubes (Figure S11 in the Supporting Information). This
demonstrates how specific ion effects may be used for post-
modification of ionic self-assembled nanostructures, in this
case, from unilamellar tubes to multilamellar rods/ribbons.
Based on the observed wall and layer dimensions within the
À
acting only weakly with water. The same is true for the PF6
ion known to be a soft anion extreme in studies of ion-specific
effects (Hofmeister series).[17] The weakly bound water is easily
À
expelled when the soft PF6 ion forms close ion pairs with the
À
ClÀ nanotubes and PF6 rods/ribbons, it became clear that the
likewise soft ATOTA-1+ bilayer surface.[18] For this neutralized
and weakly solvated surface, there will be a large hydrophobic
effect towards formation of multilayers. Further surface minimi-
zation occurs in the larger ribbons by folding into twisted
rods. For the ATOTA-1·Cl salt the situation is reverse as the
electrostatic interactions between the soft cationic bilayer and
the hard ClÀ ions cannot break the strong hydration of the ClÀ
ions. Overall, the bilayer is stabilized and adapts a tubular
structure with all anions solvated in the water phase and no
hydrophobic parts of ATOTA-1+ exposed to water. The tubes
will be further stabilized by the surrounding corona of nega-
tively charged ClÀ anions providing electrostatic repulsion to-
wards other tubes or other parts of the same tube.
first level of self-assembly common to both structures is a bilay-
er formed from the amphiphilic columnar ATOTA-1+ aggre-
gates as schematized in Figure 4B and C. This structure is very
similar to Langmuir and LB films formed by ATOTA-1·PF6 at the
air/water interface.[7–8] Comprehensive structural studies of
these films based on compression isotherms and in situ X-ray
scattering established the molecular footprint or surface area
of each ATOTA-1+ unit in the columnar aggregates to be 58 ꢃ2
(Figure 4A).[7] On the other hand, densely packed alkyl chains
each takes up an area of close to 20 ꢃ2.[7] For ATOTA-1+ with
its two n-decyl chains this corresponds to 40 ꢃ2. For simple
amphiphiles, such asymmetry in the space-filling requirements
of the hydrophilic and hydrophobic parts agree well with for-
mation of flexible bilayers.[15] The details of the next level of
self-assembly of the chloride salt towards formation of nano-
À
As was mentioned above, addition of PF6 ions to such
stable solutions of ClÀ tubes breaks this special bilayer stabili-
zation and results in formation of multilayer aggregates similar
Chem. Eur. J. 2014, 20, 1 – 5
www.chemeurj.org
3
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Communication
À
[1] a) F. J. M. Hoeben, P. Jonkheijm, E. W. Meijer, A. P. H. J. Schenning, Chem.
Rev. 2005, 105, 1491–1546; b) T. Kato, N. Mizoshita, K. Kishimoto,
Angew. Chem. 2006, 118, 44–74; Angew. Chem. Int. Ed. 2006, 45, 38–68.
[2] a) B. R. Kaafarani, Chem. Mater. 2011, 23, 378–396; b) S. Sergeyev, W.
Pisula, Y. H. Geerts, Chem. Soc. Rev. 2007, 36, 1902–1929.
to those formed directly from ATOTA-1·PF6. Thus, PF6 ions
may be applied as glue between the bilayers due to their fa-
vorable interaction with the soft ATOTA-1+ bilayer surface.
Recently, Chen et al. showed the ability to control the forma-
tion of either rigid multilayer tubular nanostructures or rigid
nanoribbons comprised of a cationic pyrene based amphiphile
by using monovalent or bivalent anions, respectively.[19] Wu
[3] a) T. Aida, E. W. Meijer, S. I. Stupp, Science 2012, 335, 813–817; b) A. C.
Coleman, J. M. Beierle, M. C. A. Stuart, B. Macia, G. Caroli, J. T. Mika, D. J.
van Dijken, J. W. Chen, W. R. Browne, B. L. Feringa, Nat. Nanotechnol.
2011, 6, 547–552; c) J. P. Hill, W. S. Jin, A. Kosaka, T. Fukushima, H. Ichi-
hara, T. Shimomura, K. Ito, T. Hashizume, N. Ishii, T. Aida, Science 2004,
304, 1481–1483; d) H. Shao, J. Seifert, N. C. Romano, M. Gao, J. J.
Helmus, C. P. Jaroniec, D. A. Modarelli, J. R. Parquette, Angew. Chem.
2010, 122, 7854–7857; Angew. Chem. Int. Ed. 2010, 49, 7688–7691;
e) D. Q. Wu, R. L. Liu, W. Pisula, X. L. Feng, K. Mullen, Angew. Chem.
2011, 123, 2843–2846; Angew. Chem. Int. Ed. 2011, 50, 2791–2794.
[4] a) W. Zhang, W. S. Jin, T. Fukushima, N. Ishii, T. Aida, Angew. Chem. 2009,
121, 4841–4844; Angew. Chem. Int. Ed. 2009, 48, 4747–4750; b) Y. Ya-
mamoto, G. X. Zhang, W. S. Jin, T. Fukushima, N. Ishii, A. Saeki, S. Seki, S.
Tagawa, T. Minari, K. Tsukagoshi, T. Aida, Proc. Natl. Acad. Sci. USA 2009,
106, 21051–21056; c) Y. N. He, Y. Yamamoto, W. S. Jin, T. Fukushima, A.
Saeki, S. Seki, N. Ishii, T. Aida, Adv. Mater. 2010, 22, 829; d) D. M. Eisele,
J. Knoester, S. Kirstein, J. P. Rabe, D. A. Vanden Bout, Nat. Nanotechnol.
2009, 4, 658–663; e) D. M. Eisele, C. W. Cone, E. A. Bloemsma, S. M.
Vlaming, C. G. F. van der Kwaak, R. J. Silbey, M. G. Bawendi, J. Knoester,
J. P. Rabe, D. A. Vanden Bout, Nat. Chem. 2012, 4, 655–662; f) D. M.
Eisele, H. V. Berlepsch, C. Bottcher, K. J. Stevenson, D. A. V. Bout, S. Kirst-
ein, J. P. Rabe, J. Am. Chem. Soc. 2010, 132, 2104; g) R. Charvet, Y. Yama-
moto, T. Sasaki, J. Kim, K. Kato, M. Takata, A. Saeki, S. Seki, T. Aida, J. Am.
Chem. Soc. 2012, 134, 2524–2527.
[5] D. Q. Wu, W. Pisula, V. Enkelmann, X. L. Feng, K. Mullen, J. Am. Chem.
Soc. 2009, 131, 9620–9621.
[6] H. Shao, M. Gao, S. H. Kim, C. P. Jaroniec, J. R. Parquette, Chem. Eur. J.
2011, 17, 12882–12885.
[7] J. B. Simonsen, K. Kjaer, P. Howes, K. Norgaard, T. Bjornholm, N. Harrit,
B. W. Laursen, Langmuir 2009, 25, 3584–3592.
[8] J. B. Simonsen, F. Westerlund, D. W. Breiby, N. Harrit, B. W. Laursen, Lang-
muir 2011, 27, 792–799.
[9] a) F. Westerlund, J. Elm, J. Lykkebo, N. Carlsson, E. Thyrhaug, B. Aker-
man, T. J. Sorensen, K. V. Mikkelsen, B. W. Laursen, Photochem. Photobiol.
Sci. 2011, 10, 1963–1973; b) B. W. Laursen, J. Reynisson, K. V. Mikkelsen,
K. Bechgaard, N. Harrit, Photochem. Photobiol. Sci. 2005, 4, 568–576.
[10] W. Brown, K. Johansson, M. Almgren, J. Phys. Chem. 1989, 93, 5888–
5894.
[11] a) S. Yagai, M. Yamauchi, A. Kobayashi, T. Karatsu, A. Kitamura, T. Ohba,
Y. Kikkawa, J. Am. Chem. Soc. 2012, 134, 18205–18208; b) L. S. Li, H. Z.
Jiang, B. W. Messmore, S. R. Bull, S. I. Stupp, Angew. Chem. 2007, 119,
5977–5980; Angew. Chem. Int. Ed. 2007, 46, 5873–5876; c) A. Gopal, M.
Hifsudheen, S. Furumi, M. Takeuchi, A. Ajayaghosh, Angew. Chem. 2012,
124, 10657–10661; Angew. Chem. Int. Ed. 2012, 51, 10505–10509.
[12] B. W. Laursen, F. C. Krebs, M. F. Nielsen, K. Bechgaard, J. B. Christensen,
N. Harrit, J. Am. Chem. Soc. 1998, 120, 12255–12263.
À
et al. have demonstrated that going from ClÀ to BF4 counter-
ions changes the self-assembly morphology of an amphiphilic
discotic p-conjugated cation in methanol from a ribbon struc-
ture to a mixture of coiled and tubular nanostructures, respec-
tively.[14] But in these cases, the morphology changes result
from packing effects in the crystalline multilayer structures re-
lated to size and valency of the anion rather than their hard/
soft property or solvent interactions.
In summary, the self-assembly of cationic amphiphilic
amino-triangulenium salts with ClÀ and PF6 counterions was
À
investigated. In both salts bilayers featuring columnar p-stacks
were formed. We demonstrated how the interaction between
these positively charged bilayers can be tuned from attractive
to repulsive by soft and hard anions, leading to rather well-de-
fined multilayer nanoribbons and nanorods, or to highly uni-
form 29 nm unilamellar nanotubes that are very flexible and
stable in aqueous solution. In particular, the combination of
a highly charged polarizable surface of the self-assembled
nanostructures (bilayers) and hard highly water coordinated
(chloride) anions turn out to be a promising recipe for stabili-
zation of discrete p-conjugated nanostructures in aqueous
media. This study empathize the great potential of ionic build-
ing blocks and electrostatic interactions for nanostructure en-
gineering via the counterions.
Experimental Section
ATOTA-1·PF6 was synthesized according to literature procedures.[7]
The new chloride salt (ATOTA-1·Cl) was synthesized analogously
from tris(2,4,6-trimethoxyphenyl)carbenium chloride.[12] Experimen-
tal details and full spectroscopic characterization are given in the
Supporting Information. Experimental details for aggregates forma-
tion, electronic absorption, emission, DLS, AFM, Cryo-TEM, and
XRD measurements are also given in Supporting Information along
with additional data/Figures S1–S11.
[13] L. Ziserman, H. Y. Lee, S. R. Raghavan, A. Mor, D. Danino, J. Am. Chem.
Soc. 2011, 133, 2511–2517.
[14] D. Q. Wu, L. J. Zhi, G. J. Bodwell, G. L. Cui, N. Tsao, K. Mullen, Angew.
Chem. 2007, 119, 5513–5516; Angew. Chem. Int. Ed. 2007, 46, 5417–
5420.
Acknowledgements
[15] J. N. Israelachvili, Intermolecular and Surface Forces, 3rd ed., Elsevier Sci-
ence, 2011.
[16] B. W. Laursen, T. J. Sorensen, J. Org. Chem. 2009, 74, 3183–3185.
[17] W. Kunz, Current Opinion in Colloid & Interface Science 2010, 15, 34–39;
Interface Science 2010, 15, 34–39.
We thank Gunnel Karlsson, Lund University, for her valuable
help with the cryo-TEM. This work was supported by the
Danish–Chinese Center for Self-Assembled Molecular Nano-
Electronics, funded by the Danish Basic Research Foundation,
and from “Center for Synthetic Biology” at Copenhagen Uni-
versity funded by the UNIK research initiative of the Danish
Ministry of Science, Technology, and Innovation.
[18] K. D. Collins, Methods 2004, 34, 300–311.
[19] Y. L. Chen, B. Zhu, Y. Han, Z. S. Bo, J. Mater. Chem. 2012, 22, 4927–4931.
Keywords: amphiphiles · nanostructures · nanotubes · self-
assembly · triangulenium salts · p-conjugation
Received: March 9, 2014
Published online on && &&, 0000
&
&
Chem. Eur. J. 2014, 20, 1 – 5
www.chemeurj.org
4
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Communication
COMMUNICATION
&
Amphiphiles
D. Shi, C. Schwall, G. Sfintes, E. Thyrhaug,
P. Hammershøj, M. Cꢀrdenas,
J. B. Simonsen, B. W. Laursen*
&& – &&
Counterions Control Whether Self-
Assembly Leads to Formation of
Stable and Well-Defined Unilamellar
Nanotubes or Nanoribbons and
Nanorods
Specific-ion effects: Self-assembly of
the amphiphilic p-conjugated carbeni-
um ion dodecyl-aminobisdimethylami-
notrioxatriangulenium hexafluoro-phos-
phate (ATOTA-1+) in aqueous solution
selectively leads to discrete and highly
stable nanotubes or nanoribbons and
nanorods, depending on the nature of
the counterion (ClÀ vs. PF6À, respective-
ly; see scheme).
Chem. Eur. J. 2014, 20, 1 – 5
www.chemeurj.org
5
ꢂ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ